Leipzig / Jena / Ilmenau. Mobile apps like Flora Incognita, which enable automated identification of wild plants, not only identify plant species, but also discover large-scale ecological models. These patterns are surprisingly similar to those derived from long-term inventory data of German flora, although they were acquired over much shorter periods of time and are influenced by user behavior. This opens up new perspectives for rapid detection of changes in biodiversity. These are the main findings of a study by a team of researchers from central Germany, recently published in Ultrasound.
With the help of artificial intelligence, today’s plant species can be classified with great precision. Smartphone apps harness this technology to allow users to easily identify plant species in the field, giving lay people access to biodiversity at their fingertips. In the context of climate change, habitat loss and land use change, these apps can serve another purpose: by gathering information on the locations of identified plant species, valuable data sets are created, potentially providing researchers with information on changing environmental conditions.
But is this information reliable – as reliable as the information provided by data collected over long periods of time? A team of researchers from the German Center for Integrative Biodiversity Research (iDiv), the Remote Sensing Center for Earth System Research (RSC4Earth) at the University of Leipzig (UL) and the Helmholtz Center for Environmental Research ( UFZ), the Max Planck Institute for Biogeochemistry (MPI-BGC) and the Technical University of Ilmenau wanted to find an answer to this question. The researchers analyzed the data collected with the Flora Incognita mobile app between 2018 and 2019 in Germany and compared it to the FlorKart database of the German Federal Agency for Nature Conservation (BfN). This database contains long-term inventory data collected by over 5,000 floristic experts over a period of over 70 years.
Mobile app uncovers macroecological patterns in Germany
The researchers report that Flora Incognita data, collected over just two years, allowed them to uncover macroecological patterns in Germany similar to those derived from long-term inventory data of German flora. The data therefore also reflected the effects of several environmental factors on the distribution of different plant species.
However, direct comparison of the two datasets revealed major differences between Flora Incognita data and long-term inventory data in areas with low human population density. “Of course, the amount of data collected in a region is highly dependent on the number of smartphone users in that region,” said the latest author, Dr Jana Wäldchen of MPI-BGC, one of the developers of the app. mobile. The differences in the data were therefore more pronounced in rural areas, with the exception of well-known tourist destinations such as Zugspitze, Germany’s highest mountain, or Amrum, an island on the North Sea coast.
User behavior also influences the plant species recorded by the mobile application. “Plant observations made with the app reflect what users see and what interests them,” said Jana Wäldchen. Common and remarkable species have been recorded more often than rare and inconspicuous species. Nevertheless, the large amount of plant observations still allow us to reconstruct familiar biogeographical patterns. For their study, the researchers had access to more than 900,000 data entries created in the first two years after the application was launched.
Automated species recognition has great potential
The study shows the potential of this type of data collection for research on biodiversity and the environment, which could soon be integrated into long-term inventory strategies. “We are convinced that automated species recognition has a much greater potential than previously thought and that it can contribute to the rapid detection of changes in biodiversity,” said first author Miguel Mahecha, professor at the ‘UL and member of iDiv. In the future, an increasing number of users of applications such as Flora Incognita could help detect and analyze ecosystem changes around the world in real time.
The Flora Incognita mobile app was developed jointly by the research groups of Dr Jana Wäldchen from MPI-BGC and the group of Professor Patrick Mäder from TU Ilmenau. This is the first plant identification application in Germany using deep neural networks (deep learning) in this context. Fueled by thousands of plant images, identified by experts, it has already made it possible to identify more than 4,800 plant species.
“When we developed Flora Incognita, we realized that there was huge potential and growing interest in improved technologies for the detection of biodiversity data. As computer scientists, we are happy to see how our technologies make an important contribution to biodiversity research, ”said co-author Patrick Mäder, professor at TU Ilmenau.